The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes untrusted data from external sources, which constitutes an indirect prompt injection surface.
Ingestion points: The skill fetches data from Jira issues via searchJiraIssuesUsingJql and Figma designs via get_design_context as specified in SKILL.md.
Boundary markers: No explicit boundary markers or warnings to ignore instructions within external data are defined.
Capability inventory: The skill has access to Bash, Write, Edit, and Agent tools, enabling it to modify the file system and execute shell commands.
Sanitization: There is no mention of sanitizing external data before it influences the agent's plan or code output.
Mitigation: The skill requires the agent to present a plan and obtain explicit user approval before implementation, which is a key security control.
[COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute development commands such as linting and type checking (tsc --noEmit). While standard for development, these commands are dynamically determined by the project environment.

code-like-me